Project Summary/Abstract Glioblastoma (GBM) is the most common malignant brain tumor across all ages. The standard of care to treat these very aggressive tumors is surgical resection followed by radiotherapy and treatment with the alkylating agent temozolomide (TMZ). However, most GBMs recur following treatment and are usually fatal. The development of new targeted therapeutics to treat these highly malignant brain tumors would have profound clinical implications. It has been shown that approximately 25% of recurrent GBMs exhibit a hypermutated phenotype and harbor mutations in mismatch repair (MMR) genes. Therefore, therapeutic strategies designed to target MMR-deficient gliomas are relevant to a large number of patients. This study aims to elucidate the contribution of the major MMR protein complexes (MutS? and MutL?) to the development of TMZ resistance in GBMs and to identify genetic dependencies that are induced by loss of MMR proteins. This project will leverage isogenic patient-derived GBM cell lines in which MMR protein members have been ablated using CRISPR-Cas9 technology. The research proposed will assess the sensitivity of these isogenic MMR-knockout GBM cell lines to TMZ over time, fueled by the hypothesis that loss of different MMR pathway members confers differential levels of TMZ resistance. To acquire a better mechanistic understanding of MMR-associated TMZ resistance, this work will investigate whether loss of specific MMR proteins induces changes in expression of other MMR pathway members, or proteins in closely-related DNA repair pathways, at both RNA and protein levels. Whole-genome sequencing of these isogenic MMR-knockout GBM cell lines in the absence and presence TMZ treatment will also be performed to systematically catalogue the landscape of mutations, copy-number variants, and rearrangements induced by loss of each MMR protein member, which will become an invaluable reference for the interpretation of signatures identified in human tumors. This project will then test the hypothesis that loss of MMR function in GBMs results in genetic dependencies that could be leveraged as synthetically lethal therapeutic targets. Genome-scale RNAi dependency data reveals that MMR-deficient colorectal cancers exhibit enriched dependencies on genes associated with homologous recombination (HR), including Rad52, RPA2, RPA3, and POLD1. This work will investigate whether these dependencies extend to the MMR-deficient glioma context. Lastly, genome- scale CRISPR-Cas9 technology will be employed to identify additional dependencies in other DNA repair pathways that could guide the development of therapeutic strategies to target these fatal brain tumors.